Neuromuscular constraints on muscle coordination during ...Inappropriate muscle activity is a source...

Clinical Neurophysiology xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Clinical Neurophysiology

journal homepage: www.elsevier .com/locate /c l inph

Neuromuscular constraints on muscle coordination during overgroundwalking in persons with chronic incomplete spinal cord injury

http://dx.doi.org/10.1016/j.clinph.2014.02.0011388-2457/� 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

⇑ Corresponding author at: Dept. of Rehabilitation Medicine, Emory University,School of Medicine, Atlanta, GA 30322, USA. Tel.: +1 404 727 3065; fax: +1 404 7124130.

E-mail address: [email protected] (R.D. Trumbower).

Please cite this article in press as: Hayes HB et al. Neuromuscular constraints on muscle coordination during overground walking in persons withincomplete spinal cord injury. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.001

Heather B. Hayes a, Stacie A. Chvatal b, Margaret A. French a, Lena H. Ting b, Randy D. Trumbower a,b,⇑a Dept. of Rehabilitation Medicine, Emory University, Atlanta, GA, USAb Dept. of Biomedical Engineering, Georgia Institute of Technology, Atlanta, GA, USA

a r t i c l e i n f o

Article history:Accepted 3 February 2014Available online xxxx

Keywords:Spinal cord injuryWalkingMuscle coordinationModulesMotor control

h i g h l i g h t s

� Persons with chronic incomplete spinal cord injury (iSCI) exhibit significant reduced muscle coordi-nation during overground walking as compared to age-matched adults.

� Neuromuscular constraints following iSCI contribute to person-specific deficits in overgroundwalking.

� Neuromuscular mechanisms underlying gait deficits may provide guidance for targeted SCIrehabilitation.

a b s t r a c t

Objective: Incomplete spinal cord injury (iSCI) disrupts motor control and limits the ability to coordinatemuscles for overground walking. Inappropriate muscle activity has been proposed as a source of clinicallyobserved walking deficits after iSCI. We hypothesized that persons with iSCI exhibit lower locomotorcomplexity compared to able-body (AB) controls as reflected by fewer motor modules, as well as, alteredmodule composition and activation.Methods: Eight persons with iSCI and eight age-matched AB controls walked overground at prescribedcadences. Electromyograms of fourteen single leg muscles were recorded. Non-negative matrix factoriza-tion was used to identify the composition and activation of motor modules, which represent groups ofconsistently co-activated muscles that accounted for 90% of variability in muscle activity.Results: Motor module number, composition, and activation were significantly altered in persons withiSCI as compared to AB controls during overground walking at self-selected cadences. However, therewas no significant difference in module number between persons with iSCI and AB controls whencadence and assistive device were matched.Conclusions: Muscle coordination during overground walking is impaired after chronic iSCI.Significance: Our results are indicative of neuromuscular constraints on muscle coordination after iSCI.Altered muscle coordination contributes to person-specific gait deficits during overground walking.� 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights

reserved.

1. Introduction

Incomplete spinal cord injury (iSCI) disrupts motor commandsto spinal locomotor circuitry and often severely limits the abilityto coordinate muscles for overground walking. While a healthy

motor system is capable of coordinating many muscles spanningmultiple joints for safe and efficient walking, this ability isimpaired following iSCI. More than 75% of persons with motorincomplete injuries regain some walking capacity (van Hedel andDietz, 2009), but many do not fully return to community walking(Field-Fote and Roach, 2011; van Hedel and Dietz, 2010). Unfortu-nately, we do not fully understand the underlying neuromuscularmechanisms that might contribute to this shortcoming nor howspecific changes in muscle co-activity impair overground walkingafter chronic iSCI.

chronic

http://dx.doi.org/10.1016/j.clinph.2014.02.001

mailto:[email protected]


http://www.sciencedirect.com/science/journal/13882457

http://www.elsevier.com/locate/clinph


2 H.B. Hayes et al. / Clinical Neurophysiology xxx (2014) xxx–xxx

Inappropriate muscle activity is a source of many of the clini-cally observed walking deficits that emerge in persons withchronic iSCI (Gorassini et al., 2009; Maegele et al., 2002). Locomo-tor training studies often target impaired muscle activity timing,agonist–antagonist joint level muscle coactivity, and electromyog-raphy (EMG) burst durations in an effort to improve walking ability(Gorassini et al., 2009; Grasso et al., 2004; Ivanenko et al., 2003,2004; Maegele et al., 2002; Visintin and Barbeau, 1994). However,these studies primarily focus on treadmill and body-weight sup-port training and not overground walking. Although treadmillwalking permits greater experimental control of walking condi-tions like speed and body-weight support, the ability to coordinatemuscles during these more constrained tasks does not necessarilytranslate to overground or community ambulation which often re-quire assistive devices such as a cane, walker, or crutches (Lee andHidler, 2008). Even though the mean kinematic trajectories aresimilar between treadmill and overground walking, overgroundwalking inherently requires greater step-to-step variability(Dingwell et al., 2001). Overground walking is a highly complexmotor task that requires flexible motor control strategies thatadapt muscle coordination to step-to-step variations in environ-mental and mechanical demands (Chvatal and Ting, 2012;Dingwell et al., 2001; Nielsen, 2003), especially compared to morecontrolled locomotor demands such as single-speed treadmillwalking (Dingwell et al., 2001).

The complexity of neuromuscular control required for over-ground walking is deficient after iSCI, resulting in numerous walk-ing deficits. For example, persons with iSCI present an inability tomodulate walking speed outside a small range of slow speeds(Pepin et al., 2003), a dependence on assistive devices (van Hedeland Dietz, 2009), and a failure to adjust to environmental perturba-tions that subsequently lead to increased falls (Brotherton et al.,2007). The extent of these walking deficits vary widely with injurylevel, severity, and the pathways damaged, making it difficult toassess the underlying neuromuscular mechanisms (van Hedeland Dietz, 2010).

To date, it is unclear to what extent inappropriate muscle coordi-nation contributes to overground walking deficits after chronic iSCI.Quantifying the contribution of altered muscle coordination is par-ticularly challenging due in part to the large number of muscles thatcontribute to overground walking. Non-negative matrix factoriza-tion (NNMF) quantifies this complexity via extraction of motor mod-ules, or groups of consistently co-activated muscles, that representthe ‘‘building blocks’’ of muscle coordination. Motor modules canbe characterized in terms of number, composition (i.e., number ofmuscles per motor module), and activation (i.e., duration and ampli-tude). Motor modules can be flexibly activated in combination toproduce a wide range of muscle coordination patterns during vari-ous motor tasks, with each module achieving a specific biomechan-ical outcome that subserves the overall biomechanical goal(Cappellini et al., 2006; Chvatal and Ting, 2012; Chvatal et al.,2011; d’Avella et al., 2011; Drew et al., 2008; Fox et al., 2013;Ivanenko et al., 2004; Neptune et al., 2009; Overduin et al., 2008;Torres-Oviedo et al., 2006). Motor modules also are useful inidentifying constraints on muscle coordination related to gaitdeficits in neurologic pathologies such as stroke, spinal cord injury,and Parkinson’s disease (Allen et al., 2013; Bowden et al., 2010;Cheung et al., 2009b; Clark et al., 2010; Fox et al., 2013; Rodriguezet al., 2013). Following hemiparetic stroke, as well as, Parkinson’sdisease, persons exhibit fewer motor modules during walking (Clarket al., 2010; Rodriguez et al., 2013). This reduction is closely relatedto limited walking speed and walking complexity. Similar findingshave been made in pediatric spinal cord injury (Fox et al., 2013),but not explored in adult spinal cord injury. Additionally, most stud-ies have focused on module number without extensive explorationof module composition or activation across the gait cycle.

Please cite this article in press as: Hayes HB et al. Neuromuscular constraints oincomplete spinal cord injury. Clin Neurophysiol (2014), http://dx.doi.org/10.1

Thus, the purpose of this study was to quantify neuromusculardeficits in muscle coordination during overground walking in per-sons with chronic iSCI. We hypothesized that overground musclecoordination is constrained by greater muscle co-activity in per-sons with iSCI as compared to age-matched (AB) controls. We pre-dict that persons with iSCI have fewer motor modules, as well as,altered composition (i.e., increased number of muscles per motormodule) and activation (i.e., increased duration) of motor modulesas compared to AB controls. We examined motor modules from 14muscles during a cadence-matched overground-walking task andrevealed that changes in motor module number, composition,and activation contribute to deficits in overground walking afteriSCI. Understanding the subject-specific neuromuscular constraintson muscle coordination is critical for effectively developing thera-pies that are more tailored to a heterogeneous population of per-sons with chronic iSCI and to the complexities of communityambulation.

2. Methods

2.1. Study population

Eight persons with iSCI (34.4 ± 3.8 years; mean ± 1 standard er-ror) and eight age-matched AB controls (34.1 ± 4.1 years) partici-pated in this study (Table 1). AB subjects also were selected tomatch gender and approximate body type. Ethical approval forthe study was received from the Emory University InstitutionalReview Board (IRB protocol STU00044670); informed consentand HIPAA authorization were obtained from all subjects prior totheir participation in accordance with the Declaration of Helsinki.

We included iSCI subjects with incomplete injuries to the spinalcord between levels C4 and T10 who were at least one year postinjury (i.e., chronic), were able to walk overground at least 10 mwith reciprocal pattern and without the assistance of another per-son, and were able to follow simple verbal, visual, and auditorycommands. We excluded subjects if they had a brain injury as de-fined from chart review, progressive SCI, other concurrent medicalcondition, and/or history of contraindications to surface electro-myography (EMG) such as adhesive allergy. We excluded AB par-ticipants if they had a concurrent medical condition and/orneurological impairments.

2.2. Clinical assessments

We recorded injury severity as well as lower extremity strengthand mobility using a set of standard clinical tests. The AmericanSpinal Injury Association Impairment Scale (AIS) was used to cate-gorize subject neurological injury level and completeness. Strengthwas assessed using the Lower Extremity Motor Score (LEMS) fromthe AIS (Marino et al., 1999). The Spinal Cord Injury FunctionalAmbulation Inventory (SCI-FAI) was used to identify clinicallyobservable gait deficits (Field-Fote et al., 2001), and the 10 m WalkTest identified the maximum walking speed (van Hedel et al.,2007). Walking tests were performed using the minimum assistivedevice possible for safe walking.

2.3. Equipment

We recorded surface EMGs from 14 muscles on the right leg(Fig. 1A), which included the tibialis anterior (TA), medial gastroc-nemius (MG), lateral gastrocnemius (LG), soleus (SO), vastus late-ralis (VL), vastus medialis (VM), rectus femoris (RF), medialhamstring (MH), lateral hamstring (LH), gluteus medius (GMED),gluteus maximus (GMAX), tensor fascia lata (TFL), sartorius (SART),and adductor magnus (ADDM). These muscles accounted for single

n muscle coordination during overground walking in persons with chronic016/j.clinph.2014.02.001


Table 1Subject characteristics.

Subject Gender Age(yrs)

AISlevel

Years sinceinjury

Assistivedevice

SCI-FAI (total,R)

LEMS (L/R,total)

Self-selected cadence(spm)

10MWT(s)

Number ofmodules

SCI1 M 28 C5 8 2C 26, 14 11/12, 23 55 14.4 7SCI2 M 30 C6 11 1C 38, 21 24/18, 42 80 5.8 3SCI3 F 28 C6 5 W 31, 18 13/22, 35 26 46.6 3SCI4 M 25 C5 8 1C 29, 20 22/19, 41 51 25.3 4SCI5 M 59 T3 24 1C 36, 20 24/23, 47 49 15.3 6SCI6 M 35 T7 5 W 21, 14 9/14, 23 26 43.4 3SCI7 M 31 C5 3 None 29, 17 19/24, 44 29 16.2 3SCI8 M 39 C7 14 2C 28, 17 13/25, 38 38a 28.9 4AB1 M 20 98 7AB2 M 37 1C 117 5AB3 F 23 W 110 5AB4 M 34 101 7AB5 M 58 1C 109 5AB6 M 31 W 110 5AB7 M 33 None 106 7AB8 M 37 2C 112 6

Abbreviations: AIS = American Spinal Cord Injury Association Impairment Scale; SCI-FAI = Spinal Cord Injury Functional Ambulation Inventory; LEMS = lower extremity motorscores; spm = steps per minute. Assistive devices are abbreviated as 1C = single cane/crutch, 2C = 2 crutches, W = walker.

a Actual mean cadences for SCI8 were slow = 32, self-selected = 38, fast = 39 while expected metronome cadences were slow = 30, self-selected = 35, fast = 40.

Fig. 1. Schematic of experimental setup and motor module analyses. (A) Schematic of surface EMG recordings from 14 right limb muscles and footswitch placement. (B)Schematic of non-negative matrix factorization reconstruction of observed EMG. Each time-invariant motor module (wi), displayed as bar plots in which each bar representsthe relative muscle contribution, is flexibly recruited at varying activation levels across time (Ci). The linear sum of the modules multiplied by their activations at each timepoint accounts for >90% of the variability in the observed EMG across all time points and muscles. Metrics describing module activations also are illustrated. The dashedhorizontal line on C1 indicates the 0.15 threshold. Cpeak is the maximum activation, Carea is the area under the curve in the region where C is above threshold, and Cduty is thepercent of the gait cycle in the region where C is above threshold.

H.B. Hayes et al. / Clinical Neurophysiology xxx (2014) xxx–xxx 3

and multi-joint actions spanning the ankle, knee, and hip and in-cluded at least one antagonistic pair at each joint, similar to previ-ous studies (Chvatal and Ting, 2012; Ivanenko et al., 2003).Standard skin preparation techniques were applied. Ag/AgCl dualsurface electrodes (model #272; Noraxon Inc, Scottsdale, AZ) wereplaced over the muscle belly parallel to fiber alignment. Four force-sensing resistor footswitches were placed on the plantar surface ofthe right foot at the heel and 1st, 3rd, and 5th metatarsal heads toidentify gait cycle events. The resulting signals were amplifiedusing Zerowire� wireless EMG technology (ZW180/R, CometaSystems, Italy), which has an input impedance of 20 M and a band-width between 10 and 500 Hz. All data were sampled at 2500 Hzwith an 18-bit data acquisition system (NI PCI-6259; NationalInstruments, Austin TX).

We measured self-selected cadences for AB and iSCI from threepasses across an instrumented GAITRite� mat (CIR Systems Inc,Clifton, NJ) in which subjects walked at a safe, comfortable speed.In subsequent data collection, subjects matched this cadence.


During data collection, footswitch voltages were analyzed using acustom algorithm in Matlab� (Mathworks Inc, USA) to compute ca-dence. If the difference between the desired and actual cadencewas greater than 5%, the trial was repeated. One subject (SCI8)was unable to match the cadences within 5% even after multipletrial repetitions; actual cadences are noted in Table 1 and these ac-tual cadences were matched by the age-matched control. Due togait deviations, the resolution of footswitch signals was not suffi-cient for toe off detection. However, average stance-to-swing tran-sitions during self-selected walking were estimated from theinstrumented mat collection as percentages of the gait cycle (AB63.0 ± 0.6%, iSCI 69.1 ± 3.8%). This enabled us to approximatestance and swing phases of the gait cycle.

2.4. Protocols

To quantify the complexity of muscle coordination across arange of overground walking speeds, we identified the number,




composition, and activation of motor modules necessary toreproduce each participant’s muscle activation patterns duringoverground walking. AB and iSCI subjects participated in nine ran-domized 20-s overground walking trials at three cadences abouttheir self-selected cadence: self-selected, slow cadence of 85% self-selected, and fast cadence of 115% self-selected (see Table 1). Wechose to evaluate walking at self-selected cadences for both groupsto understand walking coordination under conditions that weremost functionally relevant and similar to everyday communityambulation. A variety of speeds were used to acquire a rich set ofmuscle coordination patterns for exploring muscle coordinationand to represent a range of possible community walking speeds(Hof et al., 2002; Pepin et al., 2003; van Hedel et al., 2006). Duringeach of the nine trials, we instructed AB and iSCI subjects to stepto the beat of the metronome that was set according to the cadenceof that trial. Actual cadences in AB and iSCI subjects matched themetronome within 2% error. Subjects with iSCI walked used theirminimum assistive device for all trials. To control for the effects ofcadence and assistive device, a subset of the AB control subjects(n = 5; Table 1) also performed an additional nine trials at the slow,fast, and self-selected cadences of their iSCI match using thematching assistive device (ABmatch). A licensed physical therapistinstructed those subjects on appropriate use of the assistive deviceto preserve reciprocal gait. Subjects practiced for up to five minuteswith the assistive device. The cadences performed by ABmatch controlsubjects were successfully comparable to their matched subjectswith iSCI during slow (paired t-test of actual cadences walked,p = 0.51), self-selected (p = 0.10), and fast (p = 0.22) cadences.

2.5. Data processing for module analysis

We extracted motor modules from EMG recordings during eachsubject’s overground walking trials as described previously(Chvatal and Ting, 2012; Clark et al., 2010). First, each trial ofEMG recordings was high-pass filtered using a 30 Hz zero-lagfourth-order Butterworth filter and demeaned and then rectified,low-pass filtered (4 Hz, zero lag, fourth-order Butterworth filter),and down-sampled from 2500 to 20 Hz by taking the mean EMGlevels in 50 ms bins (Chvatal and Ting, 2012). Second, to allowfor comparison between subjects, we normalized data for eachsubject to maximum muscle EMG activity for a given muscle acrossall bins and all trials, such that data ranged from 0 to 1. Third, foreach dataset (i.e., AB, iSCI, and ABmatch) for each subject, nine trials(3 trials each at self-selected, slow, and fast cadences) were concat-enated into an m � n matrix where m is the number of muscles(14) and n is the number of data points (3600 samples; 20 sam-ples/s � 20 s/trial � 9 trials). Forth, before extraction, each musclewas normalized to unit variance such that each muscle’s variabilitywas equally weighted in the extraction. This normalization wasremoved after extraction (Chvatal and Ting, 2012; Torres-Oviedoand Ting, 2007). Finally, we used NNMF to extract motor modulesand corresponding activations from step to step (Chvatal and Ting,2012; Clark et al., 2010; Gizzi et al., 2011; Lee and Seung, 1999;Rodriguez et al., 2013). This technique was selected based onprevious work demonstrating that it provides robust estimates ofEMG muscle coordination with no constraints on the correlationof activations across a broad range of motor tasks (Clark et al.,2010; Rodriguez et al., 2013; Tresch et al., 2006). NNMF algorithmextracts modules such that the activity of each muscle is repre-sented as a linear summation of motor modules:

EMGr;m ¼Xn

i¼1

wi � ci

where EMGr,m is the reconstructed representation of the observedactivity (EMGo,m) for muscle m, wi is the motor module vector


describing the muscle contributions and weightings for module i,and ci is the module activation coefficient across time points formodule i. Each motor module, wi, is fixed across time, but the acti-vation coefficients can vary across time such that modules can becombined to explain a variety of muscle activation patterns. Forvisualization, module vectors were normalized to the maximummuscle contribution such that each muscle contribution rangedfrom 0 to 1.

To determine the number of motor modules required to accountfor the observed muscle coordination patterns, 1 to 14 moduleswere iteratively extracted from each subject’s data. We quantifiedgoodness of fit of the data reconstruction as the variability ac-counted for (VAF), or the uncentered correlation coefficient, whichdescribes the amount of variability in EMGo accounted for by EMGr

(Zar, 1974). To ensure robustness of module number selection, wereconstructed 10 bootstrapped datasets and computed confidenceintervals for the VAF of each module number (Cheung et al., 2009a;Sabatini, 2002). We selected module number (nmod) as the smallestnumber of modules that account for >90% VAF at the lower limit ofthe confidence interval. Module number increased if individualmuscles were not reconstructed with greater than 75% VAF andthe addition of another module increased this muscle’s fit by morethan 5%. These criteria are considered conservative to ensure good-ness of reconstruction (Chvatal and Ting, 2012; Clark et al., 2010).Finally, to establish confidence intervals on the muscle contribu-tions in each module, nmod modules were extracted from 10 boot-strapped versions of the dataset and the mean contribution of eachmuscle to each module computed.

2.6. Quantifying motor modules and activations for between andwithin group comparisons

To test the hypothesis that persons with iSCI exhibit moduleswith a greater co-activity, as evidenced by a greater number ofmuscles in each module compared to AB controls, we comparedmodule composition between subjects within and between groups(i.e., AB versus SCI). For this comparison, similarity between mod-ules was quantified by the Pearson correlation coefficient, r. Mod-ules were considered similar and, thus, aligned if r > 0.532corresponding the critical r value for 14 muscles (Chvatal et al.,2011; Zar, 1974). In the primary comparison between AB andSCI, the r-value, reported as mean ± 1 standard error, for each ABand SCI module represents the similarity to that module for AB1or the first AB to exhibit that module type. Module compositionwas quantified using two co-activity metrics, Wsum and Wmus. Wsum

measured the activity of muscles in a module and was defined asthe sum of significant muscle contributions (i.e., bar heights) in amodule. A muscle was considered significantly active within amodule if the confidence interval for that muscle did not includezero (i.e., significantly greater than zero). Wmus measured the num-ber of muscles in a module and was defined as the count of signif-icantly active muscles in a module. We predicted that both Wsum

and Wmus would be significantly greater for iSCI compared to AB.Based on our hypothesis, we predicted altered motor module

activation in persons with iSCI as compared to AB controls. Moduleactivations were divided into gait cycles using stance onsets com-puted from the footswitches. Mean activations for gait cycles atself-selected, slow, and fast cadences were computed for visualrepresentation and statistical comparisons. As illustrated inFig. 1B, we quantified module activation using three metrics: area(Carea), amplitude (Cpeak), and duration (Cduty). Carea defined theoverall activation level as the area under the mean activationcurve. To further characterize module activation shape, we quanti-fied Cpeak as the average peak of the module activations and Cduty asthe average duty cycle of the module activations. For duty cycle,modules were considered active in bins where C > 0.15, i.e., 15%




of max activity. Similarly, module composition and activation com-parisons were made between walking conditions for AB subjects(AB versus ABmatch, n = 5) and between ABmatch and iSCI formatched subjects (n = 5).

2.7. Inter-subject and intra-subject data reconstructions

To assess differences in modular organization and the flexibilityof motor modules to create a variety of muscle coordination pat-terns, modules extracted from one dataset were used to recon-struct data from another dataset (Gizzi et al., 2011; Torres-Oviedo and Ting, 2010). The VAF indicated the degree of similarity,with low VAF indicating different organizations underlying musclecoordination. We computed within-group similarity by iterativelyusing each individual AB’s modules to reconstruct the data fromall other AB subjects concatenated together and meaning theVAF. This was repeated for the iSCI group. Differences in modularorganization and flexibility between iSCI and AB groups were sim-ilarly quantified by iteratively using each individual AB’s modulesto reconstruct the full iSCI dataset. This was repeated for iSCI mod-ules reconstructing the AB dataset.

To quantify the effect of cadence and assistive device, an indi-vidual’s AB modules were used to reconstruct data from theirown ABmatch conditions (n = 5) and vice versa. Such reconstructionsalso allowed for a direct comparison of module activations becausethe modules were identical. Uncentered correlation coefficients (r)were used to compare these activations and reported as mean ± 1standard error. As above, we also compared Carea, Cpeak, and Cduty

between groups. For matched subjects, individual AB modulesreconstructed data from iSCI match (n = 5) and vice versa. Simi-larly, ABmatch modules reconstructed data from iSCI match and viceversa. These comparisons are shown in Fig. 3B.

Previous reports have shown that motor modules are robustacross from different walking speeds in able-bodied persons(Chvatal and Ting, 2012; Clark et al., 2010). To determine if thiswas evident with the cadences used in our study, we used eachsubjects’ modules to reconstruct data from their individualcadences. For all subjects and conditions, the VAF exceeded 85%,

Fig. 2. Persons with iSCI exhibit fewer motor modules than AB subjects duringoverground walking. (A) Bar plots showing the number of modules (mean ± 1standard error) required to explain muscle activity for SCI subjects (n = 8) and ABsubjects (n = 8). Asterisk indicates significant difference based on a Mann–WhitneyU test for independent samples. (B) Bars showing number of modules (mean ± 1standard error) for the subset of matched SCI (n = 5) and AB (n = 5) subjects. ABmatch

indicates the number of modules exhibited when the AB (n = 5) subjects wereconstrained to walk at the cadence and with the assistive device of their iSCI match.Asterisk indicates significant difference based on Wilcoxon Signed Rank test forrelated samples between AB and ABmatch.


with mean VAFs across all subjects of 91.9 ± 0.4% for self-selected,91.3 ± 0.6% for slow, and 91.4 ± 0.4% for fast. There was also nostatistical difference between the variability that the modulesaccounted for in the combined cadence dataset compared withvariability accounted for in individual cadence datasets(F3,60 = 0.616, p = 0.607). Additionally, modules extracted fromthe combined dataset accounted for more variance than thoseextracted from self-selected walking alone, confirming the useful-ness of varied cadence. From the models we found the modulesextracted from single cadence conditions alone represented eithera statistically similar subset of the full dataset modules or includedone to two distinct module.

2.8. Statistical analyses

All statistical analyses were performed in SPSS� 21 (IBM SPSSInc, USA). Results were considered significant at p < 0.05 and re-ported as mean ± 1 standard error. Due to the non-normality andcategorical nature of the module number dataset, we used theMann–Whitney U test for between-group comparisons (iSCI versusAB, iSCI versus ABmatch) and the Wilcoxon Signed Rank test forwithin group comparisons (AB versus ABmatch). We compared Wand C metrics using two-sample t-tests as their distributions werenot different from normal. VAF for inter- and intra-subject recon-structions were compared using one-way ANOVA with post-hocpairwise comparisons with Bonferroni corrections for multiplecomparisons. We adjusted our parametric tests depending on theLevene’s test for homogeneity of variances. Finally, linear regres-sions were used to test for significant relationships between mod-ule number and clinical assessment scores.

3. Results

3.1. Reduced number of motor modules during overground walkingafter chronic iSCI

Persons with iSCI required fewer motor modules to account fortheir muscle coordination patterns during overground walking ascompared to AB controls (Fig. 2A). AB controls exhibited 5.9 ± 0.4motor modules as compared to 4.1 ± 0.6 motor modules for per-sons with iSCI (U14 = 11.000, p = 0.024). Six of eight iSCI subjects re-quired less than 5 modules to account for the variability of musclecoordination patterns, while all AB subjects required more than 5modules, suggesting AB subjects have greater complexity of andflexibility in constructing muscle coordination during overgroundwalking.

The reduced locomotor complexity in the iSCI subjects was re-flected in inter-subject reconstructions. Modules extracted froman AB subject’s EMG muscle activity accounted for a larger per-centage (79.5 ± 1.0%) of all iSCIs’ muscle activity (F3,28 = 10.502,p < 0.001), suggesting that AB modules can be combined to createa broader range of muscle coordination patterns as compared topersons with iSCI (Fig. 3A). In contrast, modules extracted froman individual iSCI’s muscle activity accounted for a lower amount(68.1 ± 2.6%) of the variability seen across muscle activity in ABcontrols, suggesting insufficient complexity and flexibility to pro-duce AB-like muscle coordination patterns.

3.2. Altered composition and activation of motor modules emerge afteriSCI

While exact module composition was subject-specific, AB mod-ules exhibited a high degree of similarity in composition and acti-vation of modules (Fig. 4). All AB subjects exhibited modulescharacterized by the same four kinematic descriptors of



Fig. 3. Inter-subject reconstructions show that AB modules can be flexibly combined to reconstruct a wider range of muscle coordination patterns. The dashed gray lineindicates 90% VAF, the level of reconstruction achieved when modules were extracted from the data. (A) Bar plots showing the variance accounted for (VAF, mean ± 1standard error) when using one subject’s modules to reconstruct muscle activity from other subjects. Each plot represents a mean across 8 subjects. From left to right: MeanVAF for each individual AB modules reconstructing remaining muscle activity of AB subjects, individual iSCI modules reconstructing remaining SCIs’ muscle activity,individual AB modules reconstructing SCIs’ muscle activity, and individual SCI modules reconstructing muscle activity of AB subjects. (B–D) Bar plots showing the varianceaccounted for (VAF, mean ± 1 standard error) by reconstructions for the 5 subjects included in the cadence and assistive device matched subset. (B) Mean VAF of eachindividual ABmatch modules reconstructing their matched iSCI subject’s muscle activity versus individual SCI modules reconstructing their matched ABmatch subject’s muscleactivity in the cadence and assistive device condition. (C) Individual ABmatch modules reconstructing the same AB subject’s muscle activity from the self-selected conditionversus individual iSCI modules reconstructing their matched AB subject’s muscle activity. (D) Individual ABmatch modules reconstructing remaining ABmatch muscle activity inthe cadence and assistive device condition versus individual iSCI modules reconstructing the remaining iSCIs’ muscle activity. Asterisks indicate significant differencesbetween groups at p < 0.05.


overground walking. Based on the muscle composition and activa-tion profiles of AB subjects (Fig. 4), we characterized the modulesas plantar flexion (W1), hip/knee extension (W2 and/or W3), hipflexion (W4, W5, and/or W6), and eccentric braking (W7 and/orW8). 6 of 8 AB subjects also exhibited a module for hip adduction(W9). The plantar flexion module W1 was similar across all 8 ABsubjects (r = 0.95 ± 0.015). The hip flexion module (W4) was similaracross 7 AB subjects (r = 0.81 ± 0.028). The hip/knee extensionmodule (W2, r = 0.77 ± 0.056), eccentric braking module (W7,r = 0.69 ± 0.036), and hip adduction module (W9, r = 0.69 ± 0.063)were similar across 6 AB subjects. The eccentric braking module(W8) was similar across 5 AB subjects (r = 0.75 ± 0.040). Othermodules were similar across less than half of AB subjects. Due tothis similarity, modules extracted from one AB subject’s data wereable to reconstruct the remaining AB subjects’ data with81.4 ± 0.83% VAF (Fig. 3A).

Most AB modules were recruited during one or two specificphases of the gait cycle (Fig. 4, last column). During stance, oneor more of the hip/knee extension modules, composed of quadri-ceps (VL, VM, RF) and gluteals (GMAX, GMED), were first recruitedlikely to support the weight of the body followed by the plantarflexion module toward the end of stance likely for propulsion.The hip flexion modules, composed largely of TFL, SART, and RF,were recruited during stance and again during swing, with largeractivation early in the gait cycle for modules with greater RF con-tributions. The eccentric braking modules were composed largelyof the ankle dorsiflexion (TA) and hamstrings (MH, LH) and also re-cruited during both stance and swing. Early stance phase activationlikely controlled foot placement through eccentric braking of thefoot by the TA and of the shank by the hamstrings, while swingphase activation flexed and shortened the limb for forward pro-gression and foot-ground clearance. Finally, the hip adductionmodule, composed of the ADDM with or without other proximalmuscles, was likewise recruited during both phases possibly to


control limb placement during mid-stance and in preparation forheel contact during swing.

In contrast, iSCI subjects exhibited a wider range of modulecompositions, reflective of the heterogeneity inherent to iSCI. Anexample of modules and their activations for a representative sub-ject with the mean number of modules is shown in Fig. 5. Modulesextracted from iSCI subjects were each compared to the set ofmodules extracted from the AB subjects. All iSCI subjects exhibiteda plantar flexion module (W1) that was statistically similar to theAB plantar flexion module (r = 0.77 ± 0.040). The hip/knee exten-sion module (W2, r = 0.76 ± 0.048) and eccentric braking module(W8, r = 0.71 ± 0.057) were similar across 5 of 8 iSCI subjects andsimilar to their AB counterparts. All other modules were sharedby less than half of iSCI subjects or statistically distinct from allAB modules. A total of 3 iSCI modules were statistically distinctfrom all AB modules, mostly characterized by extensive co-activity.Due to reduced similarity between modules in iSCI subject, mod-ules extracted from one iSCI subject’s data reconstructed theremaining iSCI subjects’ data with 74.3 ± 2.3% VAF, significantlyless than AB subjects (Fig. 3A; F3,28 = 11.431, p = 0.001). Interest-ingly, the eccentric braking module (W7) recruited by AB for con-trolled foot placement in early swing was absent in all iSCI,contributing to foot drop or slap often observed after iSCI. Somesubjects appeared to use other TA-containing modules to provideeccentric TA control, but many lacked this control.

Module composition in subjects with iSCI modules differedfrom that of AB controls (Fig. 6). In subjects with iSCI, motor mod-ules consisted of greater muscle co-activity (Wmus) (t14 = 2.836,p = 0.013), as well as, higher contributions (Wsum) from the com-posing muscles (t14 = 2.606, p = 0.021) as compared to AB controls.This can be seen in the plantar flexion and eccentric braking mod-ules in the representative comparison shown in Fig. 6. AB modulescontained primarily muscles from one functional group, such asplantar flexors or hamstrings, whereas the iSCI modules contained



Fig. 4. Motor modules exhibited by each AB subject and representative activations. Bar plots represent the relative muscle contributions to each module, with musclesordered as indicated by the legend at the bottom. Standard deviation bars show the variation about the mean muscle contribution across 10 bootstrapped extractions. Eachcolumn represents modules from a single AB subject. Modules were considered similar and, thus, group in a row if r > 0.532 when compared to AB1 modules, the subject withthe most modules. Functional labels to the left indicate the purported mechanical function of that module or group of modules (Neptune et al., 2001). On the far right,representative activations for each module are shown as line plots from one of 2 representative subjects that exhibited that module (AB8 unless indicated as AB2). The linesshow the mean module activation across the gait cycle (5% bins) for self-selected walking trials.


a combination of multiple functional groups merged together frommultiple AB modules, similar to reports in persons with stroke(Clark et al., 2010). For example, muscles across multiple joints,such as the hamstrings and gluteals or the ankle plantar flexorsand knee extensors, were co-active in the iSCI eccentric brakingmodule and plantar flexion module. Some iSCI subjects alsoshowed antagonist muscles in a single module. In contrast, AB con-trols appeared to separate muscle contributions to motor modules,which may allow for differential control of these muscles and, thus,greater locomotor complexity.

The iSCI modules exhibited broader activation patterns duringoverground walking as compared to AB controls. Activation area(Carea, t14 = 3.539, p = 0.007) and duty cycle (Cduty, t14 = 4.243,p = 0.003) were greater for iSCI modules compared to AB modules.The plantar flexion and eccentric braking modules in particular


demonstrated this difference (Fig. 6). Although peak activation le-vel tended to be slightly higher for iSCI modules, it did not differsignificantly between the two groups (Cpeak, t14 = 1.071,p = 0.302). Broad activations of iSCI modules overlapped betweenmodules, as seen in Figs. 5 and 6, where multiple modules wereco-active across much of the gait cycle; while overlap in AB moduleactivation tended to be most active during distinct phases of thecycle.

3.3. AB subjects flexibly alter muscle coordination for iSCI matchedwalking conditions

AB control subjects walked faster overground at faster cadencesthan subjects with iSCI. Mean SCI self-selected cadence was44.3 ± 6.5 steps per minute, significantly slower than that of AB



Fig. 5. Motor modules and activations exhibited during overground walking in arepresentative subject with iSCI. Modules and activations from a representative iSCIsubject (SCI4) who exhibited 4 modules, the mean number for iSCI subjects. As inFig. 4, bar plots represent the muscle contributions to each module and line plotsshow the corresponding mean activations across the gait cycle. Modules wereassigned the same color and functional label if they were statistically similar tothose shown in Fig. 4 (r < 0.532). Gray modules indicate modules that weredissimilar from all AB modules.


(t12 = 8.779, p < 0.001). AB self-selected cadence was 107.9 ± 2.2(mean ± 1 standard error) steps per minute. Due to this between-group discrepancies in cadences, as well as, differences in assistivedevices used during overground walking, we assessed to what ex-tent slower cadences and use of assistive devices affect motormodule number, composition, and activation in AB controls. Whena subset of AB subjects (n = 5) were constrained to walk at iSCI-matched cadences using the matched assistive device, the numberof motor modules was significantly reduced from 5.9 ± 0.35 to3.4 ± 0.32 (p = 0.039; Fig. 2B). The number of modules did not differsignificantly between the subset of matched iSCI and AB subjects inthis condition (U8 = 12.000, p = 0.911; Fig. 2B). However, matchediSCI modules exhibited greater muscle co-activity as reflected byhigher muscle contributions within modules (Wsum, t8 = 2.806,p = 0.023); other metrics were not significantly different. Addition-ally, while the number of modules were not statistically different,the inability of iSCI modules to reconstruct ABmatch data and viceversa demonstrates that the modules differed in composition andare not interchangeable for creating muscle coordination patterns(Fig. 3B). Although not statistically different, ABmatch modules werebetter at reconstructing AB muscle coordination patterns com-pared to iSCI modules, despite having the same number of avail-able modules (Fig. 3C). This suggests that number is not the onlyindicator of flexibility and that AB subjects retain greater flexibilitywith a reduced number of modules compared to iSCI subjects with


the same number. Finally, ABmatch modules are significantly lesseffective at reconstructing ABmatch from the other subjects(t8 = 2.543, p = 0.035), suggesting their coordination strategies var-ies more across the cadences and assistive devices compared topersons with iSCI (Fig. 3D).

In addition to using a reduced number of modules, AB subjects(n = 5) reduced their activation of stance-phase modules, reflectingreliance on the assistive device for body weight support and pro-pulsion. For instance, two AB subjects that walked with a rollingwalker no longer exhibited a hip/knee extension module activeduring stance (Fig. 7A). When the AB self-selected modules wereused to reconstruct the ABmatch dataset, all activations decreased,particularly of weight-bearing and propulsive modules like theplantar flexion and hip/knee extension modules. Moreover, threeAB subjects that walked with cane or crutches (Table 1) at slowercadences (<50% of self-selected) no longer exhibited a plantar flex-ion module for propulsion while the higher cadence subject (>50%of self-selected) returned this module at a reduced activation level.When AB modules were used to reconstruct the ABmatch data foreither assistive device type, all module activations decreased inCarea, Cpeak, and Cduty (all t4 > 3.770 and p < 0.020); Fig. 7B illustratesthis in a representative subject (AB8). This was particularly evidentfor weight-bearing and propulsive modules like the plantar flexionand hip/knee extension modules.

3.4. Constraints on muscle coordination reflect walking deficits afteriSCI

All iSCI subjects showed limited maximum walking speeds andsignificantly lower self-selected cadences. All but one subject re-quired an assistive device to walk overground safely. The reducedmodule number, co-activity within modules, and broad inappro-priate module activation reveal neuromuscular mechanisms thatlikely constrain these aspects of overground walking ability. How-ever, the specific neuromuscular mechanisms underlying their def-icits reflected in module composition and activation appear to behighly subject-specific and heterogeneous likely due to the inher-ent heterogeneity of iSCI. As a result, the number of modules alonedid not predict self-selected cadence or maximum speed (R2 < 0.16,p > 0.33) and no linear relationship was seen between limbstrength (LEMS) or gait deviations in the SCI-FAI score (R2 < 0.09,p > 0.50). This is in contrast to previous reports in stroke andParkinson’s disease in which the number of motor modules scaleswith locomotor speed (Clark et al., 2010; Rodriguez et al., 2013).Rather, the neuromuscular mechanisms underlying reduced speedand SCI-FAI deficits appear to be subject-specific, as explored be-low in the discussion, and cannot be seen in module number alone.Nevertheless, across all participants, module number correlated towalking cadence (Fig. 8). There was a linear relation between thenumber of motor modules extracted for overground walking andcadence (r2 = 0.40; p = 0.003); subjects with slower cadences hadfewer motor modules. In contrast, subjects with essentially noimpairments (i.e., AB control subjects) and correspondingly highercadences had more motor modules.

4. Discussion

The purpose of this study was to quantify altered muscle coor-dination during overground walking in persons with chronic iSCI.In particular, we predicted that during overground walking, per-sons with iSCI have fewer motor modules, greater number of mus-cles within each motor module (composition), and alteredactivation of motor modules as compared to age-matched AB con-trols. Indeed, our results showed that persons with iSCI requiredfewer motor modules to explain the observed muscle coordination



Fig. 6. Modules differ between subject groups in number, composition, and activation. Comparison of modules and activations for a representative AB (AB6) and iSCI (SCI6)matched pair. SCI6 modules were considered similar and aligned to AB6 modules if r > 0.532. The r-values for each module and activation represent the similarity to AB#.Colors and functional labels indicate similarity to as described for Fig. 4. The gray modules were dissimilar in the Fig. 4 comparison but aligned when only compared to AB6’ship flexion module.


patterns during overground walking. Reduced module number,along with altered composition and activation within and acrossmodules, reflected constraints on muscle coordination that likelylimit safe and effective walking. These constraints are evidencedby the universal reduction in walking speed (10MWT) and self-se-lected cadence and the requirement of an assistive device in all butone iSCI. Below we discuss how motor module number, composi-tion, and activation reveal common as well as subject-specificmechanisms. The identification of neuromuscular deficits haveimplications for developing targeted rehabilitation strategies toimprove community ambulation in persons with chronic iSCI.

4.1. Constraints on muscle coordination during overground walkingafter chronic iSCI

Persons with iSCI have fewer ‘‘building blocks’’ for constructingmuscle activity, limiting the complexity of muscle coordination foroverground walking. Additionally, the co-activation within mod-ules and overlapping broad activation across modules limits theability to activate individual muscles or muscle groups to performspecific biomechanical subtasks. Therefore, the range of possiblemuscle coordination patterns and the flexibility to modulatepatterns is reduced. This lack of flexibility is evidenced by theintersubject reconstructions. While AB modules can be flexiblycombined to reconstruct other AB and iSCI muscle coordinationpatterns, iSCI modules can only reconstruct a limited range ofactivity and, thus, yield significantly poorer reconstructions ofother iSCI and AB patterns.


4.2. Adapting muscle coordination to novel task demands

The uninjured human neuromotor system can flexibly adjustmotor commands to coordinate muscles during various walkingdemands. For example, when AB subjects were challenged withthe novel task of walking at approximately 25% of their self-se-lected cadence using an assistive device, they adapted musclecoordination strategies to fit the task. AB subjects either reducedactivation of their self-selected modules and/or reduced the num-ber of modules required. Adjusting control strategies to account forbody-weight support through an assistive device eliminated theneed for modules observed during self-selected cadence condi-tions. In many cases, AB subjects eliminated hip/knee extensionor plantar flexion modules or greatly reduced their activation.The reduction in the number of modules required to walk at slowercadences with an assistive device suggests that the assistive devicemay serve as a substitute module for iSCIs, allowing iSCI subjects tocompensate through the use of upper body muscle activation andmechanical properties of the device to walk overground.

The inability of iSCI subjects to walk across a broad range ofspeeds prevents us from directly testing whether iSCI retain thisdegree of flexibility. However, their inability to modulate speedby modulating muscle coordination suggests that they lack theflexibility and locomotor complexity of the intact nervous system.While AB subjects have a greater number of modules available andthe ability to modulate activation of these modules, iSCI may belimited to just the reduced set exhibited at their self-selected ca-dences. Further, the increased co-activity (Wsum) in their modules



Fig. 7. AB subjects reduce their module number and activation when walking at matching iSCI cadence and assistive device. (A) Modules extracted from muscle coordinationpatterns during walking at self-selected cadences (112 steps per minute) for a representative subject (AB6) compared to modules extracted from the ABmatch condition ofwalking for that same subject (38 steps per minute with loftstrand crutches). Module number was reduced. (B) Modules (left) exhibited by a representative subject (AB8)walking at self-selected cadences. Module activations required to reconstruct muscle coordination during the self-selected condition (middle, 110 steps per minute) versusthe cadence and assistive device match condition (right, 26 steps per minute with rolling walker). All activations were reduced, particularly during stance. Colors andfunctional labels again indicate similarity to those shown in Fig. 4. (For interpretation of the references to color in this figure legend, the reader is referred to the web versionof this article.)


compared to AB and ABmatch conditions may prevent them fromperforming and individuating the biomechanical subtasks neces-sary for modulating speed. For example, their plantar flexion mod-ules were often compounded with hip flexors and knee extensors,which may reduce the propulsive forces produced. The plantarflexion module also was co-activated with other modules thatcould interfere further with propulsion, thereby limiting speed.Alternatively, the reduction in the number of modules used byAB at slower cadences with an assistive device could suggest thatthe reduced number of modules exhibited by iSCI subjects simplyreflects the task of a slower cadence and assistive device use.However, the inability of iSCI subjects to walk at higher speedsor without their assistive device implies that the reduced numberof modules and increased co-activity constrains task performance.This firm ceiling on their walking speed suggests that, even ifreduced module number is due to speed alone, they do not haveaccess to the additional modules required to produce the coordina-tion required for higher walking speeds. Further, previous work inable-bodied persons found that modules (also referred to as


synergies) are robust across speeds within AB subjects (Chvataland Ting, 2012; Clark et al., 2010), providing credence to the notionthat speed is not the only factor accounting for a reduction in mod-ule number when AB subjects walk slower. Most likely, the two areinextricably linked, such that the injury reduces the ability of thenervous system to produce muscle coordination, requiring iSCIsubjects to substitute an assistive device for certain biomechanicalsubtasks and to walk at slower speeds.

4.3. Limitations

Despite careful experimental design, our study has limitations.As noted above, because iSCIs cannot walk at faster speeds andforcing AB to walk at nearly half their cadence represents a noveland somewhat unnatural task, it is difficult to know for surewhether the reduced module number is an artifact of speed.However, this study does provide insight into the differences incoordination between iSCI and AB persons at both self-selectedspeeds and slow cadences, allowing us to infer the underlying



Fig. 8. Motor module number versus overground self-selected cadence in subjectswith iSCI and age-matched controls. Linear regression models were used to describethe relationship between self-selected cadence and module number for AB subjectswalking at self-selected cadence, subjects with iSCI walking at self-selectedcadence, and AB subjects walking at matched iSCI cadences and assistive devices(ABmatch). Data are pooled across all subjects. Data from AB subjects are representedwith open circles, data from iSCI subjects are represented as black circles, and datafrom ABmatch subjects are represented with gray circles. Thin dashed lines aboveand below the regression line define 95% confidence intervals. The model wassignificant for describing a relationship between self-selected cadence and motormodule number (r2 = 0.4; p < 0.003).


neuromuscular mechanisms that constrain iSCI overground walk-ing. The small sample size and heterogeneity of injury compromisethe ability to identify relationships between module number andgait deficits. While subject-specific relationships were seen, futurestudies in which subjects are stratified by injury characteristicsmay reveal a stronger correlation between modules and gait defi-cits. Additionally, while previous work on AB subjects allowed usto infer the mechanical function of these modules (Neptuneet al., 2009), how the absence or altered composition and activa-tion of modules in iSCI affects mechanical output remains uncon-firmed and deserves future investigation. Given the significantasymmetries exhibited by iSCI subjects (see Table 1), the numberof modules may likely be affected by whether the more or less im-paired limb was tested. Severe impairment in one limb may not befully reflected by an outcome measure, like the 10MWT, if signifi-cant asymmetry allows for compensation with a much strongercontralateral limb. This was certainly true of our fastest subjectiSCI2 and calls to attention the need to extend modular analysisto both limbs. Finally, subjects with a low number of modulesand high co-activity within modules may present similarly to sub-jects who fail to appropriately activate a high number of AB-likemodules. Both may display a low step rhythm score on the SCI-FAI and slow walking speed on the 10MWT, but the number ofmodules may vary significantly. Further exploration, including in-jury stratification, mechanical measures, and interlimb interac-tions, will allow us to more fully understand how neuromuscularmechanisms influence observable gait deficits and how therapiesmay be prescribed to target these deficits.

4.4. Clinical implications

The heterogeneity of spinal injuries, which vary with the neuralpathways affected and the severity of the injury, demands subject-specific analysis of the neuromuscular mechanisms that produce


walking deficits. Understanding and quantifying these mecha-nisms is vital for the development of more targeted therapies. Cur-rent clinical walking tests effectively describe observable gaitdeficits and track improvements, but are limited in their abilityto identify the underlying mechanisms and to predict adaptabilityto complex environments or perturbations (Barbeau et al., 2002;van Hedel et al., 2005). In many cases, the same observable deficitcan have multiple causes that vary between subjects. Additionally,many clinical tests rely on qualitative ordinal scales while even themost quantitative tests, such as the 10-m or 6-min walk tests, failto detect small deficits in higher functioning walkers (van Hedelet al., 2005). This limited sensitivity is evidenced by the absenceof correlation between clinical scores and the modules that serveas ‘‘building blocks’’ for muscle coordination. Our analyses alsoshowed that modular constraints are highly heterogeneous amongpersons with iSCI, as demonstrated by the low similarity amongiSCI modules and poor inter-subject reconstructions. Thus, it isnot surprising that a single metric like module number cannot fullypredict walking impairment.

Module analysis may provide a more sensitive tool for identify-ing the diversity of neuromuscular mechanisms that constrainoverground walking after iSCI. For example, the SCI-FAI identifieddeviations in step rhythm or relative time to advance the swinglimb. The two subjects who show delayed advancement, iSCI3and iSCI6, both exhibited prolonged activation of the hip/kneeextension module, suggesting that inappropriate prolongation ofthis module prevents hip flexion delaying advancement. OtherSCI-FAI deviations had multiple underlying mechanisms that var-ied depending on the subject. For example, SCI-FAI deviations instep height in which the toe fails to clear the ground were relatedto the weak activation of the ankle and hip flexion modules iniSCI1, but not in iSCI3 and iSCI6 who possessed greater TA activa-tion. Reduced step height in iSCI3 and 6 is more likely explained bythe activation of a co-activity module with hip extensors andflexors during swing phase. These co-activity modules aredissimilar from all AB modules and likely constrain hip flexionleading to failed toe clearance. Additionally, iSCI6 exhibited partic-ularly prolonged flexion module activation that may further con-tribute to an inability to dorsiflex the foot for clearance (Capadayet al., 1999). There was also variability in foot contact deviations,with 3 subject showing appropriate heel strike and 5 showing fore-foot or flat foot contact. iSCI subjects with heel strike used eccen-tric activation of modules containing TA, allowing for controlledfoot placement at stance onset, while the others showed little orno eccentric activation. This variability likely reflects differingpreservation of corticospinal tract projections to the TA, as TA re-ceives greater cortical input compared to other lower extremitymuscles particularly during early stance-phase eccentric control(Capaday et al., 1999; Perez et al., 2004; Schubert et al., 1997). Fi-nally, it should be noted the subject who is the most different fromAB in terms of number of modules and module similarity (iSCI6)walked at the slowest cadence, slowest speed, and had the lowestSCI-FAI score, suggesting that similarity to AB modules may be thebest predictor of overground walking function.

One striking finding from this study was that module numberwas correlated with self-selected cadence during overgroundwalking. This result implies that slower cadences (i.e., iSCI sub-jects) required fewer muscle coordination patterns. Our findingsare consistent with a prior study that showed persons with chronicstroke had fewer motor modules and walked slower (Clark et al.,2010). Indeed, persons with chronic iSCI walked slower and wereconstrained by fewer motor modules, as well as, abnormal co-activity between muscles in each motor module and prolonged,overlapping activation of these modules. Thus, we suspect that al-tered composition and activation of motor modules contributed toour subject-specific gait deficits (e.g., reduced speed and reliance




on assistive devices), which have been implicated as possiblemechanisms for gait deficits in other neurologic disorders such asstroke (Bowden et al., 2010) or Parkinson’s disease (Rodriguezet al., 2013). Based on these complimentary discoveries, we pro-pose that therapies directed at facilitating increased complexityof muscle coordination also may facilitate increased self-selectedwalking speed and subsequently lead to improved communityambulation in persons with iSCI.

4.5. Conclusions

This study quantified the constraints on overground musclecoordination that occur after chronic iSCI. In agreement with ourhypotheses, persons with iSCI were constrained by fewer motormodules characterized by increased co-activity and broad overlap-ping activation, both of which contribute to increased muscleco-activity. iSCI modules cannot explain the diversity of musclecoordination patterns expressed by AB controls, reflecting limitedflexibility in neuromuscular control that likely contributes toconstraints on gait speed and overground walking independentof walking aids. Further, subject-specific alterations in modulecomposition and activation reveal neuromuscular mechanismsunderlying gait deficits, potentially providing guidance for targetedand individualized rehabilitation plans.

Acknowledgements

This work was supported in part by NIH Grant K12 HD055931,DOD Grant SC090355, as well as, a Grant from the Petit Institute forBioengineering & Bioscience. The authors are very grateful to theparticipants. The authors would also like to thank Monica Dann-eman, Jillian Dean, David Gustafson, and Caitlin Manuel for theirassistance with data collection and Victoria Stahl for her assistancewith technical issues and manuscript proofing.

References

Allen JL, Kautz SA, Neptune RR. The influence of merged muscle excitation moduleson post-stroke hemiparetic walking performance. Clin Biomech2013;28:697–704.

Barbeau H, Fung J, Leroux A, Ladouceur M. A review of the adaptability and recoveryof locomotion after spinal cord injury. Prog Brain Res 2002;137:9–25.

Bowden MG, Clark DJ, Kautz SA. Evaluation of abnormal synergy patternspoststroke: relationship of the Fugl-Meyer assessment to hemipareticlocomotion. Neurorehabil Neural Repair 2010;24:328–37.

Brotherton SS, Krause JS, Nietert PJ. Falls in individuals with incomplete spinal cordinjury. Spinal Cord 2007;45:37–40.

Capaday C, Lavoie BA, Barbeau H, Schneider C, Bonnard M. Studies on thecorticospinal control of human walking. I. Responses to focal transcranialmagnetic stimulation of the motor cortex. J Neurophysiol 1999;81:129–39.

Cappellini G, Ivanenko YP, Poppele RE, Lacquaniti F. Motor patterns in humanwalking and running. J Neurophysiol 2006;95:3426–37.

Cheung VCK, d’Avella A, Bizzi E. Adjustments of motor pattern for loadcompensation via modulated activations of muscle synergies during naturalbehaviors. J Neurophysiol 2009a;101:1235–57.

Cheung VCK, Piron L, Agostini M, Silvoni S, Turolla A, Bizzi E. Stability of musclesynergies for voluntary actions after cortical stroke in humans. Proc Natl AcadSci USA 2009b;106:19563–8.

Chvatal SA, Ting LH. Voluntary and reactive recruitment of locomotor musclesynergies during perturbed walking. J Neurosci 2012;32:12237–50.

Chvatal SA, Torres-Oviedo G, Safavynia SA, Ting LH. Common muscle synergies forcontrol of center of mass and force in nonstepping and stepping posturalbehaviors. J Neurophysiol 2011;106:999–1015.

Clark DJ, Ting LH, Zajac FE, Neptune RR, Kautz SA. Merging of healthy motormodules predicts reduced locomotor performance and muscle coordinationcomplexity post-stroke. J Neurophysiol 2010;103:844–57.

d’Avella A, Portone A, Lacquaniti F. Superposition and modulation of musclesynergies for reaching in response to a change in target location. J Neurophysiol2011;106:2796–812.

Dingwell JB, Cusumano JP, Cavanagh PR, Sternad D. Local dynamic stability versuskinematic variability of continuous overground and treadmill walking. JBiomech Eng 2001;123:27–32.

Drew T, Kalaska J, Krouchev N. Muscle synergies during locomotion in the cat: amodel for motor cortex control. J Physiol 2008;586:1239–45.


Field-Fote EC, Fluet GG, Schafer SD, Schneider EM, Smith R, Downey PA, et al. Thespinal cord injury functional ambulation inventory (sci-fai). J Rehabil Med2001;33:177–81.

Field-Fote EC, Roach KE. Influence of a locomotor training approach on walkingspeed and distance in people with chronic spinal cord injury: a randomizedclinical trial. Phys Ther 2011;91:48–60.

Fox EJ, Tester NJ, Kautz SA, Howland DR, Clark DJ, Garvan C, et al. Modular control ofvaried locomotor tasks in children with incomplete spinal cord injuries. JNeurophysiol 2013;110:1415–25.

Gizzi L, Nielsen JF, Felici F, Ivanenko YP, Farina D. Impulses of activation but notmotor modules are preserved in the locomotion of subacute stroke patients. JNeurophysiol 2011;106:202–10.

Gorassini MA, Norton JA, Nevett-Duchcherer J, Roy FD, Yang JF. Changes inlocomotor muscle activity after treadmill training in subjects with incompletespinal cord injury. J Neurophysiol 2009;101:969–79.

Grasso R, Ivanenko YP, Zago M, Molinari M, Scivoletto G, Castellano V, et al.Distributed plasticity of locomotor pattern generators in spinal cord injuredpatients. Brain 2004;127:1019–34.

Hof AL, Elzinga H, Grimmius W, Halbertsma JP. Speed dependence of averaged EMGprofiles in walking. Gait Posture 2002;16:78–86.

Ivanenko YP, Grasso R, Zago M, Molinari Marco, Scivoletto G, Castellano V, et al.Temporal components of the motor patterns expressed by the human spinalcord reflect foot kinematics. J Neurophysiol 2003;90:3555–65.

Ivanenko YP, Poppele RE, Lacquaniti F. Five basic muscle activation patterns accountfor muscle activity during human locomotion. J Physiol 2004;556:267–82.

Lee DD, Seung HS. Learning the parts of objects by non-negative matrixfactorization. Nature 1999;401:788–91.

Lee SJ, Hidler J. Biomechanics of overground vs. Treadmill walking in healthyindividuals. J Appl Physiol 2008;104:747–55.

Maegele M, Muller S, Wernig A, Edgerton VR, Harkema SJ. Recruitment of spinalmotor pools during voluntary movements versus stepping after human spinalcord injury. J Neurotrauma 2002;19:1217–29.

Marino RJ, Ditunno Jr JF, Donovan WH, Maynard Jr F. Neurologic recovery aftertraumatic spinal cord injury: data from the model spinal cord injury systems.Arch Phys Med Rehabil 1999;80:1391–6.

Neptune RR, Clark DJ, Kautz SA. Modular control of human walking: a simulationstudy. J Biomech 2009;42:1282–7.

Neptune RR, Kautz SA, Zajac FE. Contributions of the individual ankle plantar flexorsto support, forward progression and swing initiation during walking. J Biomech2001;34:1387–98.

Nielsen JB. How we walk: central control of muscle activity during human walking.Neuroscientist 2003;9:195–204.

Overduin SA, d’Avella A, Roh J, Bizzi E. Modulation of muscle synergy recruitment inprimate grasping. J Neurosci 2008;28:880–92.

Pepin A, Ladouceur M, Barbeau H. Treadmill walking in incomplete spinal-cord-injured subjects: 2. Factors limiting the maximal speed. Spinal Cord2003;41:271–9.

Perez MA, Lungholt BK, Nyborg K, Nielsen JB. Motor skill training induces changes inthe excitability of the leg cortical area in healthy humans. Exp Brain Res2004;159:197–205.

Rodriguez KL, Roemmich RT, Cam B, Fregly BJ, Hass CJ. Persons with Parkinson’sdisease exhibit decreased neuromuscular complexity during gait. ClinNeurophysiol 2013;124:1390–7.

Sabatini AM. Identification of neuromuscular synergies in natural upper-armmovements. Biol Cybern 2002;86:253–62.

Schubert M, Curt A, Jensen L, Dietz V. Corticospinal input in human gait: modulationof magnetically evoked motor responses. Exp Brain Res 1997;115:234–46.

Torres-Oviedo G, Macpherson JM, Ting LH. Muscle synergy organization is robustacross a variety of postural perturbations. J Neurophysiol 2006;96:1530–46.

Torres-Oviedo G, Ting LH. Muscle synergies characterizing human posturalresponses. J Neurophysiol 2007;98:2144–56.

Torres-Oviedo G, Ting LH. Subject-specific muscle synergies in human balancecontrol are consistent across different biomechanical contexts. J Neurophysiol2010;103:3084–98.

Tresch MC, Cheung VCK, d’Avella A. Matrix factorixation algorithms for theidentification of muscle synergies: evaluation on simulated and experimentaldata sets. J Neurophysiol 2006;95:2199–212.

van Hedel HJ, Dietz V, Curt A. Assessment of walking speed and distance in subjectswith an incomplete spinal cord injury. Neurorehabil Neural Repair2007;21:295–301.

van Hedel HJ, Wirz M, Dietz V. Assessing walking ability in subjects with spinal cordinjury: validity and reliability of 3 walking tests. Arch Phys Med Rehabil.2005;86:190–6.

van Hedel HJA, Dietz V. Rehabilitation of locomotion after spinal cord injury. RestorNeurol Neurosci 2010;28:123–34.

van Hedel HJA, Dietz VGroup EMSoHSCIS. Walking during daily life can be validlyand responsively assessed in subjects with a spinal cord injury. NeurorehabilNeural Repair 2009;23:117–24.

van Hedel HJA, Tomatis L, Müller R. Modulation of leg muscle activity and gaitkinematics by walking speed and bodyweight unloading. Gait Posture2006;24:35–45.

Visintin M, Barbeau H. The effects of parallel bars, body weight support and speedon the modulation of the locomotor pattern of spastic paretic gait. Apreliminary communication. Paraplegia 1994;32:540–53.

Zar JH. Biostatistical analysis. Englewood Cliffs, NJ: Prentice-Hall; 1974.


http://refhub.elsevier.com/S1388-2457(14)00066-2/h0005



























































































































Date post:	26-Sep-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

Neuromuscular constraints on muscle coordination during ...Inappropriate muscle activity is a source...

Documents